Abstract

Soccer video processing and analysis to find critical events  such  as  occurrences  of  goal  event  have  been  one  of  the important issues and topics of active researches in recent years. In this paper, a new role-based framework is proposed for goal event detection in which the semantic structure of soccer game is used.  Usually  after  a  goal  scene,  the  audiences’  and  reporters’ sound  intensity  is  increased,  ball  is  sent  back  to  the  center  and the  camera  may:  zoom  on  Player,  show  audiences’  delighting, repeat the goal scene or display a combination of them. Thus, the occurrence  of  goal  event  will  be  detectable  by  analysis  of sequences of above roles. The proposed framework in this paper consists of  four  main procedures: 1-  detection of game’s critical events by using audio channel, 2-  detection of shot boundary and shots classification, 3-  selection of candidate events according to the  type  of  shot  and  existence  of  goalmouth  in  the  shot,  4 -detection  of  restarting  the  game  from  the  center  of  the  field.  A new  method  for  shot  classification  is  also  presented  in  this framework.  Finally,  by  applying  the  proposed  method  it  was shown that the goal events detection has a good accuracy and the percentage of detection failure is also very low.